Adverse effects of noise exposure on health

Transcription

1 TNO report Adverse effects of noise exposure on health a state of the art summary Division Public Health Wassenaarseweg 56 P.O. Box CE Leiden The Netherlands T F Info-VGZ@pg.tno.nl Date 27 September 2001 Authors Y de Kluizenaar W Passchier-Vermeer HME Miedema 2001 TNO

2 TNO report / 22 Authors: Y de Kluizenaar W Passchier-Vermeer HME Miedema Project number Deze uitgave is te bestellen door het overmaken van ƒ 21,- (incl. BTW) op postbankrekeningnummer ten name van TNO PG te Leiden onder vermelding van bestelnummer

3 TNO report / 22 Abstract This document is prepared by order of the Department of Technology Assessment of the University of Stuttgart, within the framework of UNITE (UNIfication of accounts and marginal costs for Transport Efficiency). Within the UNITE program, exposureeffect relationships are needed for a monetary valuation of transportation noise. The aim of this project is to give an overview of the state-of-the-art of noise effects research and to provide an overview of exposure-effect relationships. Available exposure-effect relationships are given for the effects of transportation noise on various health aspects in the living environment, in offices and other otherwise quiet working environments and recreational areas. Some of these relationships are well established, others are of a preliminary nature. In addition, an estimate of the importance and occurrence of different types of effects in the Netherlands is given. With respect to disturbances during outdoors recreation, results of a number of investigations carried out in National Parks in the United States are reviewed. However, it is uncertain whether these relationships can be applied in Europe, where conditions are different.

4 TNO report / 22 Contents 1 Introduction 5 2 Acoustical aspects 6 3 Effects of noise exposure on health Introduction Description of health effects 9 4 Noise exposure-effect relationships overview for the living environment Hearing impairment Stress related health effects Psychosocial effects Sleep disturbance Cognitive effects on school children Speech interference Annoyance in recreational areas 17 5 Conclusion 19 Appendices A Appendix

5 TNO report / 22 1 Introduction This document is prepared by order of the Department of Technology Assessment of the University of Stuttgart, within the framework of UNITE (UNIfication of accounts and marginal costs for Transport Efficiency). Within the UNITE program, exposure-effect relationships are needed for a monetary valuation of transportation noise. This study aims to give an overview of the state-of-the-art of noise effects research. The main focus is to produce a list describing the currently available exposure-effect relationships for the effects of transportation noise on health. In addition, the document gives an estimate of the importance and occurrence of different types of effects in the Netherlands. When addressing the issue of adverse effects on health, the term health needs to be defined first. Health effect is defined by the WHO document Guidelines for community noise (1999) as: an adverse effect of noise is defined as a change in the morphology and physiology of an organism that results in impairment of functional capacity, or an impairment of capacity to compensate for additional stress, or increases the susceptibility of an organism to the harmful effects of other environmental influences. Health effects are diverse. When characterizing effects both acute (temporary) and long lasting effects can be distinguished, some of the latter being permanent. There is sufficient scientific evidence that exposure to noise can lead to adverse health effects. This insight is not new. A large number of effects of noise exposure on health and quality of life where already known or at least hypothesized in the 1960 s. Since then, a large number of epidemiological studies have been carried out that have confirmed and refined earlier insights. Currently a large variety of scientific information is available on a wide range of noise effects. In general, exposure specified by the environment in which persons are exposed includes four categories: 1 exposure in the living environment (including schools); 2 exposure in the working environment; 3 exposure during leisure time; 4 exposure during transportation. In the living environment the main sources of noise include road traffic, rail traffic, air traffic, industry and neighbourhood activities. Noise sources in the working environment depend largely on the kind of activities and may be generated both within and outside the workplace. The main sources of noise in the living environment may also have an impact on office workers and other people during working hours if the sounds enter their otherwise quiet work environment. Exposure during leisure time includes exposure of (mainly young) people to pop music (e.g. concerts, nightclubs, portable CD-players and walk-mans), exposure during games and sports (shooting) and exposure of children to noisy toys. People may also be exposed to sound during recreation outdoors (nature areas, campsites, lakes, etc.). This document will focus on the effects of exposure to transportation noise in the living environment, in offices and other otherwise quiet working environments and recreational areas. The effect of noise within factories is outside the scope of this project and is therefore not considered here. In chapter 4 exposure-effect relationships are given. Some of these relationships are well established, others are of a preliminary nature.

6 TNO report / 22 2 Acoustical aspects Noise can be described by various noise metrics. The human hearing organ is not equally sensitive to sounds of different frequencies, therefore the sound pressure level (L) is A-weighted, and expressed as db(a). Important noise metrics are: 1 L Aeq,T : The equivalent sound level over a period of time: L Aeq,T = 10 lg ( 1 / T T 10 L(t)/10 ) dt. 2 L dn : The day-night level. This metric is the equivalent sound level over a 24-hour period, with the sound levels during nighttime (22 7 h) increased by 10 db(a). 3 L den : The day-evening-night level. This metric is the equivalent sound level over a 24- hour period, with the sound levels increased by 5 db(a) and 10 db(a) during the evening (19 23 h) and during nighttime (23 7 h) respectively. 4 SEL: Sound Exposure Level. A metric used to describe single noise events. This metric can be described as the equivalent sound level during the event, normalized to a period of 1 second. In this document L den is chosen as the general noise metric for the 24 hour period, and L Aeq, night with usually a duration of 8 hours (23 07 h) is used for the nighttime exposure. L den will most likely be selected as the standard metric by the European Union. SEL is used to describe the exposure to single noise events. Exposure-effect functions are presented using these metrics to describe the noise exposure. Usually to evaluate an environmental noise situation, values of these noise metrics are assessed outdoors. Where appropriate, indoors noise levels can be assessed by taking into account the sound insulation of the dwelling. For road traffic noise ISO 717-1:1996 specifies a noise spectrum given in the upper part of Table 1. If these relative octave band sound pressure levels are exponentially added, the result is 0 db. In the Netherlands for living rooms, large and small bedrooms the sound insulation is given at the lower part of the table (Reubsaet, 1998). The sound insulation has been defined as the difference between the sound pressure level measured outside the dwelling at the facade (without reflections) and measured indoors. Taking into account the ISO noise spectrum, the Netherlands sound insulation, and the A- weighting, the sound insulation for living rooms and large bedrooms is 28 db(a), for smaller bedrooms 21 db(a). In the Netherlands also the sound insulation of bedrooms for aircraft noise has been specified as 21 db(a). Table 1 Road traffic octave band spectrum (ISO 717-1:1996) (upper part of the table) and sound insulation in octave bands for the Netherlands situation (lower part of the table). Midfrequency octave band in hertz SPL per octave in db ISO spectrum Sound insulation in db for: Living room Large bedroom Smaller bedroom

7 TNO report / 22 3 Effects of noise exposure on health 3.1 Introduction The effects of exposure to noise on health can be divided in the following groups: 1 hearing impairment; 2 stress related health effects (hypertension, cardiovascular diseases, effects on birth weight); 3 psychosocial effects (annoyance, effects on psychosocial well-being); 4 sleep disturbance; 5 effects on performance; 6 speech interference. These effects have been examined in various epidemiological studies. In 1994, the Committee on Noise and Health, an international committee of the Health Council of the Netherlands (HCN), assessed the health effects of environmental and occupational noise exposure. Noise induced health effects with sufficient evidence for a causal relationship between exposure and effect as described by the Health Council of the Netherlands (1994) are: hearing loss, annoyance, hypertension, ischaemic heart disease, sleep disturbance and effects on cognitive performance of children in schools. Table 2 gives an overview of the state-of-the-art of noise effects research. This table is based on Table 2 of the 1994 Health Council report, and has been updated since 1994 (see also Passchier- Vermeer and Passchier, 2000). Scientific evidence for different groups of effects has been rated in terms of sufficient, limited, inadequate, or lacking, respectively. Also an observation threshold has been defined, where sufficient scientific evidence was considered available. The observation threshold for an effect is defined as the lowest noise exposure level at which on average the effect was observed in epidemiological studies. In the update recent reviews have been taken into consideration (Institute for Environment and Health, 1997; Morrell, et al., 1997; Shaw, 1996; Job, 1996; Carter and Job, 1998). Changes include the metric used to describe the noise exposure observation threshold for certain effects, such as hypertension and ischaemic heart disease. Secondly, the observation threshold for awakening by a single noise event has been lowered by 5 db(a), and finally an observation threshold for sleep pattern changes and speech interference has been added.

8 TNO report / 22 Situation b Table 2 Long term and possible long term effects of exposure to noise and classification of the evidence for a causal relationship between noise and effect. The last three columns contain information on the observation threshold of an effect for which the causal relationship with noise exposure is judged to be sufficient. Effect Classification of evidence a Observation threshold Reference Metric Value in db(a) Indoors/ Outdoors c Hearing impairment Sufficient Env L Aeq,24h 70 1 in ISO, 1999 Hypertension Sufficient Env d L den 70 2 out HCN, 1994 Ischaemic heart disease Sufficient Env L den 70 3 out HCN, 1994 Biochemical effects Limited Env Immune effects Limited Env Birth weight Limited Env air Prenatal effects Lack Env Psychiatric disorders Limited Env air Annoyance Sufficient Env L den 42 e,4 out Miedema and Oudshoorn, 2001 Psychosocial well-being Limited Env Performance Sufficient School L Aeq,school 70 5 out Passchier-Vermeer, 2000 Speech interference Sufficient Env, School SEL 65 6 in / out Miedema and Passchier- Vermeer, 1999 Sleep disturbance, changes in: sleep pattern Sufficient Sleep awakening Sufficient Sleep SEL 55 7 in HCN, 1997 sleep stages Sufficient Sleep SEL 35 8 in HCN, 1994 subjective sleep quality Sufficient Sleep L Aeq,night 40 9 out HCN, 1997 heart rate Sufficient Sleep SEL in HCN, 1994 hormone levels Limited Sleep immune system Inadequate Sleep mood next day Sufficient Sleep L Aeq,night >60 11 out HCN, 1994 performance next day Limited Sleep a) Classification of evidence of causal relationship between noise and health. b) Env = living environment, Sleep = sleep period time, School = exposure of children at school. c) Value relates to indoor or outdoor noise assessment. d) On average the differences between L den and L dn are only marginal. By analysis over situations with road, railway or noise, the difference between L den and L dn appeared to be less than 0.5 db(a). In specific situations however, the difference substantially larger. e) The observation threshold is about 12 db(a) lower for environmental impulse noise Exposure-effect relationships are given in Table 3 (if available).

9 TNO report / Description of health effects Health effects have been described by the Health Council of the Netherlands, Committee on noise and health (1994) and by Passchier-Vermeer and Passchier (2000), and will be briefly summarized here. Hearing impairment Hearing impairment is defined as an increase in the hearing threshold level. A hearing handicap is defined as the disadvantage imposed by hearing impairment sufficiently severe to affect ones personal efficiency in the activities of daily living. This disadvantage is usually expressed in terms of understanding conventional speech in low level ambient noise (International Standard; ISO, 1999). Stress related health effects Exposure to noise above a certain threshold can act as a stressor. Reactions to a stressor can be psychological, behavioural and somatic in nature. Cardiovascular effects such as ischaemic heart disease and hypertension belong in that third category. Ischaemic heart disease includes angina pectoris and myocardial infarction. Hypertension has been defined by the World Health Organization as a systolic blood pressure of at least 160 mmhg (this value corresponds to approximately 0.13 kpa), and/or a diastolic blood pressure of at least 95 mmhg. Psychosocial effects Psychosocial effects due to environmental noise that have been studied in epidemiological investigations include annoyance, psychosocial well-being, and psychiatric hospitalization. The main psychosocial effect from exposure to occupational noise observed in epidemiological investigations is annoyance. Annoyance is a feeling of resentment, displeasure, discomfort, dissatisfaction, or offence when noise interferes with someone s thoughts, feelings or actual activities. Annoyance in populations is evaluated using questionnaires. In this report exposure-effect relationships for noise annoyance are included as derived at TNO from a large international archive of noise annoyance studies (Miedema and Vos, 1998; Miedema and Oudshoorn, 2001). These relationships are described in a report (Miedema and Oudshoorn, 2001) which was prepared on behalf of the EU/DG Environment for the Working Group Dose/Effects which has the task to provide relationships between noise exposure and noise effects, for the preparation of an EU directive on environmental noise. Annoyance questions in different international studies do not use the same number of response categories. This number varies from 3 to 11. In order to obtain comparable annoyance measures for all different studies, all sets of response categories have been translated into a scale ranging from 0 to 100, assuming that a set of annoyance categories divides the range from 0 to 100 in equally spaced intervals. In the definition of annoyance a cutoff point is chosen of 72 for the percentage highly annoyed (%HA), 50 for the percentage annoyed (%A) and 28 for the percentage (at least) a little annoyed (%LA). Sleep disturbance Sleep is a recovery process that is essential for humans to function properly. From the sleep EEG, two distinct phases of sleep can be distinguished: the NREM sleep and REM (rapid eye movement) sleep. NREM sleep covers four stages. Stages 1 and 2 are called light sleep, and stages

10 TNO report / 22 3 and 4 are called deep sleep (slow wave sleep: SWS). In stages 1 and 2 the transition to REM sleep or awakening occurs. In general, body- and brain restoration mainly occur during the SWS and the REM sleep respectively. During sleep, adults have sleep cycles of about 90 minutes, in which REM and NREM sleep occur alternating. Adverse health effects are expected from chronic noise-induced interference with sleep, as it impairs the functions of sleep with respect to brain and body restoration (Horne, 1990; Carter, 1998). In addition to the physiological aspects of a noise-induced reduction of sleep quality, nighttime noise exposure of sufficient intensity is also related to subjectively experienced sleep quality (Passchier-Vermeer et al., 1998). Also, reduced sleep quality interferes with daytime functioning, adversely affecting mood next day and possibly cognitive performance of adults. Sleep quality can be quantified by measures of subjective, behavioural, or physiological responses. To quantify subjective sleep quality sleep logs, diaries or questionnaires are used. The most commonly applied behavioural and physiological methods used to quantify sleep quality are EEG recordings, actimetry and pressing a button. Effects on performance There is sufficient evidence from laboratory experiments that the presence of noise can significantly impair cognitive performance. Noise can induce learned helplessness, increase arousal, alter the choice of task strategy, and decrease attention to the task. Noise may also affect social performance, mask speech and other sound signals, impair communication and distract attention from relevant social clues. Exposure of schoolchildren to noise can adversely affect their performing of cognitive tasks. Speech interference Noisy events, which increase the ambient noise level, may cause interference of conversation either by disturbance of the speaker (the speaker will have to speak louder) or by masking speech resulting in a reduced comfort of the listener and a decrease in sentences perceived.

11 TNO report / 22 4 Noise exposure-effect relationships overview for the living environment During a technical meeting on noise issues within the framework of UNITE a subset of health effects due to noise in the living, working environment and recreational areas was selected for further assessment. The selected environmental noise-induced effects are hypertension, ischaemic heart disease, annoyance, performance of school children, subjective sleep quality, speech interference, speech interference during working hours in offices and other otherwise quiet working environments and disturbance due to exposure to transportation noise during outdoors recreation in recreational areas. This report will focus on these selected issues, and their exposureeffect relationships (if available). In Table 3 exposure-effect relationships are listed. Table 3 Exposure-effect relationships for exposure to various noise sources in the living environment. Preliminary (worst case) relationships for speech interference due to outdoors traffic noise in offices are also given. Exposure-effect relations listed for effects 1 to 6 are based on a 24 hours exposure, exposure-effect relations listed for effects 7-11 are based on exposure during the sleep period time only. Exposure-effect relationships relate to adult populations. Effect Exposure-effect relation Reference 1 Hearing impairment DHTL T<10 = (u + v * lg 10) (lg (T + 1) / lg 11) * (L - L 0) 2. DHTL T>10 = (u + v * lg T) * (L - L 0) 2. Where: DHTL: median hearing loss T: exposure time (years) u,v,l o: constants dependent on the frequency of hearing impairment L: L Aeq, 8h Note: These relationships are defined bij ISO (1999) for exposure to noise in the working environment. For environmental noise exposure, the constant L o is decreased by 5 db(a) and taken over a 24 hours period ISO, Hypertension RR = * L den for L den > 70 db(a) for road traffic and aircraft noise. No data for railway noise exposure is available (see text). Preliminary relationship where RR is the average relative risk. 3 Ischaemic heart disease RR = * L den For L den > 70 db(a) for road traffic and aircraft noise. No data for railway noise exposure is available (see text). Preliminary relationship where RR is the average relative risk. 4 Annoyance Aircraft (little annoyed, annoyed, highly annoyed): %LA = * 10-4 (L den 32) * 10-2 (L den 32) (L den 32); %A = * 10-6 (L den 37) * 10-2 (L den 37) * (L den 37); %HA = * 10-5 (L den 42) * 10-2 (L den 42 ) (L den 42). 5 Performance Unknown Road traffic: %LA = * 10-4 (L den 32) * 10-2 (L den 32) (L den 32); %A = * 10-4 (L den 37) * 10-2 (L den 37) * (L den 37); %HA = * 10-4 (L den 42) * 10-2 (L den 42) (L den 42). Rail: %LA = * 10-4 (L den 32) * 10-2 (L den 32) (L den 32); %A = * 10-4 (L den 37) * 10-3 (L den 37) * (L den 37); %HA = *10-4 (L den 42) * 10-3 (L den 42) (L den 42). Miedema and Oudshoorn, Speech interference f(sel) = 0.05 (SEL 65) f(sel) = 1 f(sel) = 0 for 65 db(a) < SEL < 85 db(a); for SEL > 85 db(a); for SEL < 65 db(a). Miedema and Passchier- Vermeer, 1999 Where:

12 TNO report / 22 f(sel) : probability of speech interference in relation to SEL Worst case situation: N si = * 10 L/ * 10 (L 15)/10 (L 25)/ * 10 HCN, 1997 Where: L: L Aeq,07-23h (L assessed outdoors at the most exposed facade of the dwelling without reflections). N si: Number of speech interruptions a year. Worst case situation offices: N si = * 10 L/10 Where: N si: Number of speech interruptions in offices due to environmental noise during a year (with 250 working days of 8 working hours) and with a probability of conversation equal to L: L Aeq,8h 7 Awakening P w,n = 0.18 Σ i (SEL ind,i 55) HCN, Sleep stages Unknown Where: P w,i : The average percentage of awakenings due to a noise event during sleep period time. SEL ind,i: The SEL value determined in the bedroom for noise event i. Worst case situation: Number of awakenings a year = * 10 L/10 Where: L: L Aeq,night in the bedroom with night 8 hours, e.g. from to hours. 9 Subjective sleep quality Aircraft (highly sleep disturbance annoyed): %HS = 0.48 (L Aeq,23-07h 32.6) (preliminary relationship); Road traffic: %HS = 0.62 (L Aeq,23-07h 43.2) (preliminary relationship); Rail: %HS = 0.32 (L Aeq,23-07h 40.0 ) (preliminary relationship). L Aeq, 23-07h is assessed outside at the most exposed façade. HCN, Heart rate Unknown 11 Mood next day Unknown 4.1 Hearing impairment ISO 1999 gives a method to estimate noise-induced hearing impairment in populations exposed to continuous, intermittent or impulse noises during working hours. Exposure-effect relationships show that noise-induced hearing impairment predominantly occurs in the higher frequency range of Hz, with largest effects observed at 4000 Hz. With increasing equivalent sound level and increasing exposure time, effects can also occur at lower frequencies (Passchier-Vermeer and Passchier, 2000). In Passchier-Vermeer (1993b) it has been argued that environmental noise exposure in the Netherlands will most likely cause no hearing impairment (HCN, 1994). Possibly, in megacities environmental noise exposure is so high that it may cause noise induced hearing impairment.

13 TNO report / Stress related health effects There are large individual differences in susceptibility. Also, people can intervene in their own situations, for instance by moving from noisy surroundings to quieter places. This may result in noise proof populations exposed to the higher noise levels (TNO and RIVM, 1998). Not withstanding these complications, conclusions on the relationship between noise exposure and cardiovascular disease appear possible from meta-analyses of the available epidemiological data on aircraft and road traffic noise. For railway noise no data is available (HCN, 1994). Taking into account that railway noise is less annoying and less sleep disturbing than road traffic noise, for railway noise the exposure-effect relationship may assumed to be shifted relative to road traffic noise by 5 db(a) to higher noise levels. A large number of laboratory experiments, reviewed by Passchier-Vermeer (1993a), have shown noise induced temporal changes in the cardiovascular system. Epidemiological environmental noise studies on changes in blood studies pressure and increased risk for ischaemic heart disease in adults are limited mainly to the effects of road traffic noise, with the exception of a Dutch study on the effects of aircraft noise (Knipschild, 1977). Several studies have observed that noise exposure can result in an increase in the percentage of people suffering from hypertension (As reviewed by Passchier-Vermeer and Passchier, 2000). Noise induced increase in occurrence of ischaemic heart disease and hypertension are described below. a) Ischaemic heart disease The relative risk of ischaemic heart disease for persons living in areas with road or traffic noise starts to increase at equivalent sound levels over 24 hours above 70 db(a) measured at the facade outside the dwelling (HCN, 1994). Below this level, no noise-induced effect is to be expected (e.g. Babisch et al., 1998). Data in recent publications on cardiovascular effects from exposure to noise are not inconsistent with previous findings (e.g. Ising et al., 1997). For the purpose of this project exposure-effect relationships have been derived for myocardial infarction. Possibly noise induced increase in number of hospital admissions for myocardial infarction is correlated with the increase in actual deaths as a result of this disease. The noise induced increase in number of deaths per year, may be estimated by: n id = (n d / n ha ) * n iha Where: n id = noise induced increase in number of deaths (myocardial infarction ) per year; n ha = total number of hospital admissions (myocardial infarction ) per year; n d = total number of deaths (myocardial infarction ) per year; n iha = noise induced increase in number of hospital admissions per year. Estimates of occurrence of noise induced ischaemic heart disease (myocardial infarction) and hypertension in the Netherlands are given in Appendix I. b) Hypertension and blood pressure The relative risk of hypertension for adults living in areas with road or aircraft noise starts to increase at equivalent sound levels over 24 hours above 70 db(a) measured at the facade outside the dwelling (HCN, 1994). Housing features such as double glazing and sound insulation, and personal habits (opening windows, moving to the quiet side of the house etc.) affect the actual noise exposure (HCN, 1994). It is unknown how these features affect the risk of hypertension.

14 TNO report / 22 There is also an effect of noise on blood pressure of children (Passchier-Vermeer, 2000). The effects of exposure to noise on systolic and diastolic blood pressure in children has been examined in several studies (Cohen et al., 1980; Regecova and Kellerova, 1995; Evans et al., 1998; Hygge et al., 1996). These studies show a significant noise-induced increase in systolic and diastolic blood pressure in children. In some studies it cannot be excluded that social class has confounded the results (see also Lercher et al., 1998). In a German study (Evans et al., 1998; Hygge et al., 1996), schoolchildren were examined in the years Munich airport moved from one location to another. One location was situated close to the old airport and the other close to the new airport. The cross-sectional part of the study showed a marginally significant higher systolic blood pressure in children highly noise exposed at school. Children were matched on socio-economic characteristics. In the morning stress hormones were examined. Overnight resting levels of epinephrine and norepinephrine were significantly higher in the children exposed to aircraft noise at the old Munich airport in comparison to control groups. After the move of the airport levels of epinephrine and norepinephrine rose significantly among children living under the flight paths of the new airport. 4.3 Psychosocial effects Environmental noise exposure significantly contributes to noise annoyance. Important effectmodifying factors are noise sensitivity, fear of the noise source, and a feeling that the noise could be avoided. These effect modifying factors have been identified in multivariate analyses of population data (Guski, 1999; Job, 1999; Miedema and Vos, 1999; Stallen, 1999). It is not possible to predict noise annoyance on an individual basis because of the large variety of (partly unknown) endogeneous and exogeneous characteristics that affect annoyance. However, relationships between noise annoyance and noise exposure have been established on a population level and the effects of several modifying factors have been quantified. 4.4 Sleep disturbance Although in the 1990s several field studies were started or completed (Passchier-Vermeer et al., 1998; Fidell et al., 1998; Griefahn et al., 1996) there is still and urgent need for a tested model on the relationships between the adverse effects on sleep, and environmental night-time noise exposure, in which causal and modifying factors and their mutual relations are specified. Subjective sleep quality Preliminary relationships between self-reported sleep disturbance and nighttime ( h) equivalent sound level measured at the most exposed facade outside of the dwelling have been derived from the same data archive of noise effect surveys as was used to specify annoyance relationships (HCN, 1997). In Table 2 preliminary relationships are given for the percentage of highly sleep-disturbance annoyed respondents (%HS). The percentage highly annoyed is determined using a cut-off point at 72 on a scale from 0 ( not at all annoyed ) to 100 ( very much annoyed ). These linear regression equations currently form the best estimate for quantification. Awakening Awakenings may be assessed by a number of different methods, including EEG recordings, subjects pressing a button after awakening, subjects reporting awakening the following morning and recordings of physical motility activity by actimeters. Relationships have been derived for noise and noise-induced awakenings. Awakenings are here defined as awakenings determined by EEG recordings. The relationship between the percentage of awakenings in an adult person and the SEL of an isolated noise event is given in Table 2. This (preliminary) relationship represents

15 TNO report / 22 the current level of knowledge; further investigation is necessary to obtain more definitive relationships (HCN, 1997). In a worst case situation the number of noise-induced awakenings per year in a healthy person used to sleep in the specific situation is given by: Number of awakenings during a year = * 10 L/10, where: L = L Aeq,night with night 8 hours, e.g. from to hours. L is assessed in the bedroom. E.g., if L = 30 db(a) the number of noise-induced awakenings is in the worst case situation equal to 96 times a year. Awakening during sleeping time and subjectively experienced sleep quality are closely related (Passchier et al., 1998). To avoid that effects of nighttime noise are counted twice it is recommended that exposure-effect relationships for subjective sleep quality be used to describe effects on sleep. 4.5 Cognitive effects on school children Some of the adverse effects of noise on children may be closely related to effects of noise on their caretakers. Studies show significant interruptions and lost teaching time in schools with high traffic noise levels. Most studies on the psychological effects of noise on children focussed on aspects of cognition (e.g. Cohen et al., 1980; Hygge et al., 1996; Haines et al., 1998). The bestdocumented noise effect is the effect on reading acquisition. There are fewer studies of noise effects on other aspects of cognitive processing, such as long term memory, attention, and motivation of children. Some studies have shown that children highly exposed to environmental noise for prolonged periods of time are less motivated when placed in situations where task performance is dependent on persistence. Further research into the mechanisms and contributing factors relating to noise exposure and adverse cognitive effects on children is needed to quantify effects. 4.6 Speech interference Currently, exposure-effect relationships for speech interference in normal living conditions have not yet been extensively tested. However, on the basis of laboratory data, a preliminary bestestimate exposure-effect curve is given for vocal effort at 1 m from the speaker and the required level of speech for adequate reception (Heusden et al., 1979; Plomp, 1986). At an ambient level of 55 db(a) the speech level is just about sufficient to have an effect on the reception threshold. The speech reception threshold of a person is defined as the level of speech (db(a)) at which the person is able to correctly reproduce 50% of the sentences can be heard by the receiver. It is assumed that the risk of speech interference increases linearly from 0 at an ambient level of 55 db(a) to 1 at an ambient level of 75 db(a) for an average population (Houtgast, personal communication). The relationship depends on specific circumstances, including individual variations in speech level, speech quality and willingness to raise speech loudness during higher ambient noise level periods. It is important to note that these relationships are estimated for a speech level measured at 1 m from the speaker. In practice, the distance between speaker and listener will often be larger, resulting in reduced speech reception. In addition, the A-weighted level only gives an indication of speech interference; reliability of the use of this metric depends on the difference in sound spectrum of ambient noise and speech. Furthermore, in laboratory experiments constant ambient noise levels were used. In normal living circumstances ambient noise characteristics will be

16 TNO report / 22 different. Duration of the elevated background noise level also plays an important role, as people will behave differently depending on the duration of an intruding noise. During a short spell of noise, people will tend to stop speaking until the noise level drops. During longer periods of elevated noise levels, people will tend to stop their conversation completely, or change their location. Relationships have been derived for speech interference and noise levels during isolated noise events (expressed in SEL) assuming an average duration of such events of 10 seconds (see also Miedema and Passchier-Vermeer, 1999), using the following tentative relationships for the probability of speech interruption during a noise event and SEL of the noise event: f(sel) = 0.05(SEL 65) if SEL is between 65 en 85 db(a); f(sel) = 1 if SEL is larger than 85 db(a); f(sel) = 0 if SEL is smaller than 65 db(a). Conversations usually only take place during a part of the time. Let z be the probability that a person is having a conversation. The probability of speech interruption is then: f(sel) = 0.05z(SEL 65) if SEL is between 65 en 85 db(a); f(sel) = z if SEL is larger than 85 db(a); f(sel) = 0 if SEL is smaller than 65 db(a). By a method similar to the one applied in the calculation of noise-induced awakenings, the number of speech interruptions in the worst case situation are calculated as a function of L. The calculations are more complicated, mainly because it has to be taken into account that people speak at various locations inside and outside the dwelling, with different shielding from outside noises. Let p G be the probability of being indoors with windows closed and assume a sound insulation of 25 db(a), let p O be the probability of being indoors with windows partly opened and assume the sound insulation to be 15 db(a), and assume that the rest of the time is spent outdoors (without sound insulation). The starting point is a person who is always at home. It is also assumed that conversation is independent of the location of the speaker and listener. The maximal incidence of noise-induced speech interruptions is here assumed to be the sum of the expected maxima for each of three situations: outside, inside with closed windows and inside with partly opened windows. To be able to calculate the maximal incidence, values for z (probability of conversation), p G (probability of being indoors with windows closed) and p O (probability of being indoors with windows partly opened) have to be chosen. The probability of being outdoors, is 1 (p G + p O ). Assume that z = 0.10, p G = 0.75 and p O = 0.22 for the calculations in the following example. The probability of being outdoors between 06 and 22 hours is therefore equal to 0.03 (which is about half an hour a day). It is further assumed that speech only takes place during the day and evening, and not during the night. Worst case situation: N si = * 10 L/ * 10 (L 15)/ * 10 (Miedema and Passchier-Vermeer, 1999). (L 25)/10 Where: L = L Aeq,07-23h assessed outdoors at the facade of the dwelling. N si = number of speech interruptions in a year.

17 TNO report / 22 It should be stressed that in situations with a specified L-value, but with another noise event pattern than the worst case situation, the number of speech interruptions is smaller and may even be zero. For the purpose of this project the number of speech interruptions in offices due to environmental noise during a year (with 250 working days of 8 working hours) is estimated. For the worst case situation this number of speech interruptions is, with a probability of conversation equal to 0.10, equal to * 10 L/10, with L = L Aeq,8h assessed outdoors at the facade of the office. 4.7 Annoyance in recreational areas Under the National Park Overflight Act (1987) a series of investigations into the effects of aircraft passages to nature areas has been carried out in a number of areas in the United States. These investigations resulted in several reports, including Fidell, Silvati and Pearsons (1990); Fidell et al. (1992); Sneddon, Silvati, Pearsons and Fidell, (1991); Tabachnick, Fidell and Silvati, (1991); Tabachnick, Howe and Fidell, (1992); Anderson et al., (1993). The report of Andersen et al., (1993) gives a number of exposure-effect curves (Figure 1 and Figure 2). However, it is uncertain whether exposure-effect curves derived for National Parks in the United States can be used to quantify effects in other areas, such as nature areas in Europe. No exposure-effect curves have yet been derived for recreational areas in Europe. Miedema (1995), Jong (1998), Gerretsen (1998), and Borst (1999) developed a method to evaluate noise exposure in recreational areas. This method has until now not been applied on such a scale that reliable exposure-effect relations could be established.

18 TNO report / 22 Figure 1 The percentage annoyed (A 40 ) as a function of the L Aeq during a visit. At the left a worksheet is given that can be used to take into account the influence on annoyance of variables other than the L Aeq. Firstly an adapted L Aeq value is determined, subsequently a percentage annoyed can be read from the figure (source: Anderson et al., 1993). Figure 2 The percentage of visitors for which the natural quiet is being disturbed (NQ 40 ) as a function of L Aeq during a visit. At the left a worksheet is given that can be used to take into account the influence on this disturbance of a number of variables other than the L Aeq. Firstly an adapted L Aeq value is determined, subsequently a percentage annoyed can be read from the figure (source: Anderson et al., 1993).

19 TNO report / 22 5 Conclusion This report provides an overview of exposure-effect relationships for the effects of transportation noise on health in the living environment and preliminary relationships for speech interference from outdoors traffic noise in offices and other otherwise quiet working environments. Where exposure-effect relationships were not well documented, but current knowledge allowed a preliminary relationship, a best estimate has been given.

20 TNO report / 22 References ANDERSON GS, HORONJEFF RD, MENGE CW, et al. Dose-response relations derived from data collected at Grand Canyon, Haleakala and Hawaii Vulcanoes international parks. Lexington (Mass.): Harris Miller &Hanson, HMMH Report no , NPOA Report no BABISCH W, ISING H, GALLACHER JEJ, SWEETNAM PM, ELWOOD PC. The Caerphilly and Speedwell studies, 10 year followup. in: Carter NL, Job RFS, eds. Proceedings of Noise as a Public Health Problem (Vol. 1 & 2), Sydney, Australia: Noise Effects : BORST HC. Implementatie rekenmethoden Stiltegebieden. Leiden: TNO-PG, Publ. nr CARTER NL, JOB RFS, eds. Proceedings of Noise as a Public Health Problem (Vol. 1 & 2), Sydney, Australia: Noise Effects : CARTER NL. Cardiovascular response to environmental noise during sleep. In: Carter NL, Job RFS, eds. Proceedings of noise as a Public Health Problem, vol. 1 & 2. Sydney: Noise Effects : COHEN S, EVANS GW, KRANTZ DS, STOKOLS D. Physiological, motivational and cognitive effects of aircraft noise on children: moving from the laboratory to the field. Am Psychol 1980;35: JONG RG de. Beoordelingsmethode Stiltegebieden Deelrapport Belevingsonderzoek. Leiden: TNO-PG, JONG RG de, STEENBEKKERS JHM, VOS H. Hinder en andere zelf-gerapporteerde effecten van milieuverontreiniging in Nederland: inventarisatie verstoringen Leiden: TNO Prevention and Health, Pub. Nr EVANS GW, BULLINGER M, HYGGE S. Chronic noise exposure and physiological response: a prospective study of children living under environmental stress. Psychol Sci 1998;9: FIDELL S, SILVATI L, PEARSONS KS. Acoustic measurement of sonic booms and ambient sound levels in the Sellway-Bitterroot Wilderness area. Canoga Park: BNN Systems and Technology, BNN report no. 7196, NPOA Report no FIDELL S, SILVATI L, TABACHNICK B, et al. Short term effects of aircraft overflights on outdoor recreationists in three wildernisses. Canoga Park CA: BNN Systems and Technology, BNN report no. 7502, NPOA Report no FIDELL S, HOWE R, TABACHNICK B, et al. Field studies of habituation to change in nighttime aircraft noise and of sleep motility measurement methods. Report no Canogan Park (A), CA: BBN Systems and Technologies Corporation (1998).

21 TNO report / 22 FRANSSSEN EAM, JONG RG de LEBRET, E, MIEDEMA HME, POLL HFPM van, VOS H, WALDA IC, WIECHEN CMAG van. Hinder, slaapverstoring, gezondheids- en belevingsaspecten in de regio Schiphol, resultaten van een vragenlijstonderzoek [Annoyance, sleep disturbance, health aspects and the perception of the living environment around Schiphol airport, results of a questionnaire survey. In Dutch]. Leiden; Bilthoven: TNO Preventie en Gezondheid; Rijksinstituut voor Volksgezondheid en Milieu, TNO-PG Publ. nr , RIVM rapportnr GERRETSEN E. Beoordelingsmethode Stiltegebieden-Deelrapport Rekenmethode. Delft: TPD- HAG-RPT , GRIEFAHN B, MEHNERT P, MOEHLER U, SCHUEMER-KOHRS A, SCHUEMER R. Design of a field study on the effects of railway noise and road traffic noise. In: Proceedings of Inter- Noise 96, Aug. 1996, St. Albans, UK: Insitute of Acoustics, 1996: GUSKI R. Personal and social variables as co-determinants of noise annoyance. Noise & Health 1999; 3: HAINES MM, STANSFELD SA, JOB RFS, BERGLUND B. Chronic aircraft noise exposure and child cognitive performance and stress. In: Carter NL, Job RFS, eds. Proceedings of Noise as a Public Health Problem (Vol. 1 & 2), Sydney, Australia: Noise Effects : HEALTH COUNCIL OF THE NETHERLANDS (HCN) Committee on Noise and Health. Noise and health [Geluid en gezondheid]. The Hague: Health Council of the Netherlands, Publication no. 1994/15 E ). HCN. Committee on Uniform environmental noise exposure metric. Assessing noise exposure for public health purposes. The Hague: Health Council of the Netherlands; publication no. 1997/23 E, 112p (1997). HEUSDEN E VAN, PLOMP R, POLS LCW. Effect of ambient noise on the vocal output and the preferred listening level of conversation speech. App Acoustics 12: 31-9 (1979). HORNE J. Why we sleep: the functions of sleep in humans and other mammals. New York: Oxford University Press, HYGGE S, EVANS G, BULLINGER M. The Munich airport noise study: cognitive effects on children from before to after the change over of airports. In: Proceedings of Inter-Noise 96, Aug. 1996, St. Albans, UK: Insitute of Acoustics (1996). INSTITUTE FOR ENVIRONMENT AND HEALTH. The Non-Auditory Effects of Noise. Report R10. Leicester, UK (1997). INTERNATIONAL ORGANIZATION FOR STANDARDIZATION. Acoustics Determination of occupational noise exposure and estimation of noise-induced hearing impairment. International Standard ISO Geneva, Switzerland: International Organization for Standardization (1999). ISING H, BABISCH W, KRUPPA B, LINDTHAMMER A, WIENS D. Subjective work noise: a major risk factor in myocardial infarction. Soz Praventivmed 1997;42: JOB RFS. Noise sensitivity as a factor influencing human reaction to noise. Noise & Health 1999;3:

22 TNO report / 22 JOB RFS. The influence of subjective reactions to noise on health effects of noise. Environ Int 1996;22: KNIPSCHILD PV. Medical effects of aircraft noise: community cardiovascular survey. Int Arch Occup Environ Health 1977;40: LERCHER P, STANSFELD SA, THOMPSON SJ. Non-auditory health effects of noise: review of the period. in: Carter NL, Job RFS. eds. Proceedings of Noise as a Public Health Problem (Vol. 1 & 2), Sydney, Australia: Noise Effects : MIEDEMA HME, OUDSHOORN CGM. Annoyance from transportation noise: Relationships with exposure metrics DNL and DENL and their confidence intervals. Environ Health Perspect 2001;109(4): MIEDEMA HME, PASSCHIER-VERMEER W. Beoordeling van geluidpieken in de woonomgeving. Leiden: TNO Prevention and Health. Publ.nr MIEDEMA HME, VOS H. Demographic and attitudinal factors that modify annoyance from transportation noise. J Acoust Soc Am 1999;105: MIEDEMA HME, VOS H. Exposure response relationships for transportation noise. J Acoust Soc Am 1998;104(6): MIEDEMA HME. De beoordeling van geluid in milieubeschermingsgebieden. Leiden: TNO Preventie en Gezondheid, Publ. nr MORRELL S, TAYLOR R, LYLE D. A review of health effects of aircraft noise. Aust NZ J Publ Health 1997; 21: PASSCHIER-VERMEER W. Noise and health: review. The Hague: Health Council of the Netherlands, 1993a.. Publication no. A93/02 E. PASSCHIER-VERMEER W. Noise-induced hearing lost from daily occupational noise exposure: extrapolations to other exposure patterns and other populations. In Vallet, M, ed. Proceedings 6th International Congress on Noise and Public Health. Volume 3. Nice: INRETS, 1993, (1993b). PASSCHIER-VERMEER W, JONG RG de, MIEDEMA, HME.Geluid en gezondheid. Schattingen gezondheidseffecten door vliegtuiglawaai rond Schiphol; tweede versie. Leiden: NIPG-TNO, Publ. nr PASSCHIER-VERMEER W. Noise and health of children. Leiden: TNO Prevention and Health, Publ. nr PASSCHIER-VERMEER W, VOS H, GILS K van, MIEDEMA HME, et al. Aircraft noise and sleep disturbance. Pilot study. Leiden: TNO Prevention and Health Publ. nr PASSCHIER-VERMEER WP, PASSCHIER WF. Noise exposure and public health. Environ Health Perspect 2000;108(1):

23 TNO report / 22 PLOMP R. A signal-to-noise ratio model for the speech reception threshold of hearing impaired. J Speech Hearing Res 1986;29: REGECOVA V, KELLEROVA E. Effects of urban noise pollution on blood pressure and heart rate in preschool children. J Hypertens 1995;13: REUBSAET L. De geluidwering van de gevel bij lage frequenties. Eindhoven: Technische Universiteit Eindhoven, Faculteit Bouwkunde, FAGO-rapport A SNEDDON M, SILVATI L, PEARSONS KS, FIDELL S. Measurement and analysis of the indigenous sound environment of coniferous forests. Canoga Park CA: BBN Systems and Technology, BNN Report no. 7210, NPOA Report no SHAW E. Noise environments and the effect of community noise exposure. Noise Control Eng J 1996; 44: STALLEN PJM. A theoretical framework for environmental noise annoyance. Noise & Health 1999;3: TABACHNICK B, FIDELL S, SILVATI L. Intermediate term effects of aircraft overflights on outdoor recreationists in twelve wildernesses. Canoga Park CA: BNN Systems and Technology, BNN Report no TABACHNICK B, HOWE R, FIDELL S. Estimation of aircraft overflight exposure in national parks and forest service wildernesses. Canoga Park CA: BNN Systems and Technology, BNN Report no. 7259, NPOA Report no WORLD HEALTH ORGANIZATION. Guidelines for community noise. Geneva: WHO, 1999.

24 TNO report Appendix A.1/2 A Appendix Noise exposure in the Netherlands and it s effects. The Health Council of the Netherlands (1994) has estimated the extent of occurrence of noiseinduced health effects for the Netherlands (HCN, 1994). Estimates are subject to some uncertainty due to several factors, including the complexity of exposure-effect relations, inaccuracies in the data for the various noise exposures, intervening and confounding variables such as age, gender and stressors other than noise, combinations of exposures, effects of noise reducing measures etc. It must be noted that estimates are given for 1994, the present situation may be slightly different. Demographic data used for the Netherlands used in the report are: Number of inhabitants: about 15 million Of these 25.8 % aged under 20 years and 74.2% (11.1 million) aged 20 years and over. Number of dwellings: 5.8 million Average number of inhabitants: 2.5 persons per dwelling Table 4 describes the estimated number of people that are subject to noise-induced health effects in the Netherlands in The table is adapted from Table 3 of the Health Council of the Netherlands, Committee on Noise and Health (1994). Table 4. Estimated number of persons in the Netherlands in 1993 with noise-induced health effects presented in order of magnitude classes and as percentages of the population. Numbers are given for the adult population of the Netherlands (persons aged 20 years and over). Exposure situation and source Effect Class a Percentage of population (%) hypertension ischaemic heart disease b noise-induced hearing loss - decreased sleep quality c 7? > 10 road traffic severe annoyance 7 > 10 civil air traffic severe annoyance military air traffic severe annoyance 7 > 10 d rail traffic severe annoyance industries severe annoyance combination of traffic severe annoyance other sources in the severe annoyance 7 > 10 neighbourhood a Classes are specified as: 1 none 2 < 100 persons persons persons persons persons 7 > 1 million persons b Number per year c Person at risk; the number of persons with effects could not be estimated.